GBrowse Setup

This repository contains setup instructions, maintenance scripts and links to data for setting up a GBrowse instance on Ubuntu.

Installation

From a bare Ubuntu 14 image Gbrowse can be installed very easily. To install everything

	sudo apt-get install gbrowse gbrowse-calign gbrowse-data libbio-samtools-perl apache2

Then configure Apache to run cgi scripts

	sudo ln -s /etc/apache2/mods-available/cgid.load /etc/apache2/mods-enabled/
	sudo ln -s /etc/apache2/mods-available/cgid.conf /etc/apache2/mods-enabled/
	sudo service apache2 restart

And for speed enable fast-cgi

	sudo apt-get install libapache2-mod-fastcgi

After this is done you should be able to access the tutorial by browsing to http://hostname/gbrowse2

Under this setup the key directories are little different from those mentioned in tutorials

Original	New	Purpose
/var/www/gbrowse2	/usr/share/gbrowse/htdocs	html files
/var/lib/gbrowse2/databases	/var/lib/gbrowse/databases	databases
/etc/gbrowse2	/etc/gbrowse	gbrowse conf

So that we can easily admin the server we change the ownership of the gbrowse conf and data directories. In the original setup these are owned by root. By keeping the group unchanged these files will still be accessible to apache.

	sudo chown -R ubuntu /var/lib/gbrowse
	sudo chown -R ubuntu /etc/gbrowse

Configuring user auth

The GBrowse tutorial on Authentication provides the main details on how to do this but some crucial details are missing.

The most important is that we need to fix an issue with the ubuntu install which does not set a value for wwwuser in the config. This can be fixed as follows;

	sudo perl -e 'use GBrowse::ConfigData;GBrowse::ConfigData->set_config(wwwuser => "www-data");GBrowse::ConfigData->write;'

Then create the users database

	gbrowse_metadb_config

Check that the db has permissions for www-data. If not change its group and ownership to www-data

Then create a new user

	sudo gbrowse_create_account iracooke

Finally, in order to restrict access to a particular database place this at the top

[GENERAL]
restrict = require valid-user

Configuring databases

Properly Format gff files

An essential thing is to take heed of this note from the manual

It is very important to have a full-length entry (such as the one for ctgA) for each reference sequence mentioned in the first column of the GFF3 file. However, the reference sequence can have any source and type you choose. Commonly used types are "clone", "chromosome" and "contig."

Since programs like Augustus don't include these entries by default it will be necessary to add them.

Now create the sqlite file by ingesting a gff3 as follows. For a large file this can take a long time (several hours)

	bp_seqfeature_load -c -a DBI::SQLite -d 'genes.sqlite' soft_coral_151022.UTRon.rename.gff3

In order to enable tracks relating to the genome sequence itself we also need a db for that

	bp_seqfeature_load -c -a DBI::SQLite -d 'scaffolds.sqlite' scaffold8.fa

In the conf file specify the sqlite adaptor and the actual sqlite file for genes and scaffolds

[genes:database]
db_adaptor    = Bio::DB::SeqFeature::Store
db_args       = -adaptor DBI::SQLite
              -dsn /var/lib/gbrowse/databases/lobophytum_pauciflorum/genes.sqlite

[scaffolds:database]
db_adaptor    = Bio::DB::SeqFeature::Store
db_args       = -adaptor DBI::SQLite
               -dsn /var/lib/gbrowse/databases/lobophytum_pauciflorum/scaffolds.sqlite

BigWig support

sudo apt-get install build-essential libpng12-dev libssl-dev


wget wget http://hgdownload.soe.ucsc.edu/admin/exe/userApps.v310.src.tgz
tar -zxvf userApps.v310.src.tgz
cd userApps/kent/src/

# in inc/common.mk replace `CFLAGS=` line with `CFLAGS=-fPIC` 


make
cd ..
 
sudo -i
export MACHTYPE=x86_64
export KENT_SRC=`pwd`
cpan Bio::DB::BigWig

Logging

For basic logging see the apache error log at /var/log/apache2/error.log

bp_seqfeature_load -c -f -a DBI::SQLite -d ~/gbrowse/FH.sqlite your-assembly.fasta your-maker.gff

Blat Plugin

First install the blat executables

	cd /usr/local/bin
	sudo wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/blat/blat
	sudo wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/blat/gfClient
	sudo wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/blat/gfServer
	sudo wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/faToTwoBit
	sudo chmod a+x blat gfClient gfServer faToTwoBit

Index the genome

	cd /var/lib/gbrowse/databases/lobophytum_pauciflorum/
	faToTwoBit soft-coral_genome-assembly151022.fa soft-coral_genome-assembly151022.2bit
	mv soft-coral_genome-assembly151022.2bit /var/lib/gbrowse/twobit/
	cd /var/lib/gbrowse/twobit/

	nohup gfServer -log=gfServer.log start localhost 17781 *.2bit &

iracooke/README.md